Customer Segmentation Machine Learning

1 ) Customer Spending Profile

Highlighted in yellow:

Target Spending: Hoping close to Customer Left Year Group <=2

Historical Spending: Use as baseline, ideally not less than before

Approximately:

55,984 X Assume Response Rate X $1,045 + 2,030 X Assume Response Rate X $599 = Expected Revenue

Expected Revenue Details could be using QUANTILE_GRP.

2 ) Rencency, Frequency, Monetary Distribution

Rencency: Time since last purchase

Frequency: Store Visit

Monetary: Spending Amount

3 ) Customer Orders Distribution

Remarks: Orders actually mean store visit.

4 ) Customer Segment

4.1 Median Recency

4.2 Median Frequency

4.3 Median Monetary

5 ) Target Customer Waterfall

6 ) Contact List

Highlighted in yellow:

Total No. of Customers available to contact.

Historical Spending: Use as baseline, ideally not less than before

Approximately:

Before

55,984 X Assume Response Rate X $1,045 + 2,030 X Assume Response Rate X $599 = Expected Revenue

After

46,988 X Assume Response Rate X $1,018 + 1,684 X Assume Response Rate X $599 = Expected Revenue

Expected Revenue Details could be using QUANTILE_GRP.